DIY Video Projector Part II

[woneill]

Hi Forrest,

Check out your alignments and light path.

The light coming from your first fresnel (assuming it is a singlet) should be parallel. The second fresnel should be as close to the LCD as possible - check out which way round you have it. If it is close, and it is the right way around, there should be no distortion.

If your image is too big, then you might have an issue with focal length - even if the fresnel & objective are matched (see prev posts), if the focal lengths are too short, you will get mega zoom.

You might want to try a concave lens close after your objective to reduce zooming a little.

Bill.

[Mario007]

You do realize that forrest is using a lcd projector and not a panel right? It is an older infocus and from what I've read it only does 640*480 res which is still pretty good. The replacement bulbs for it should be fairly cheap as it just uses a 400w halogen bulb. Although replacing it with a small metal halide would be cool.

[ap0the0sis]

Has anyone thought of using the reflector side off a automotive fog light as their reflector? Just think of the times you get blined by them when you drive down the street. Now, Imagine it with a metal halide bulb in it.

I've been looking at these reflectors and they seem like they would work well since they are very reflective and they are available anywhere from 3" to 8" across (baja style) or even larger.

My SHARPVISION projector uses a 6.4" screen (800x600) and the reflector from the metal halide buld assembly is no larger than 3" across and the lens the light shines through is no larger than 2" so I dont think you need to use huge bucket style reflectors.


just a thought

ap0the0sis

[eebasist]

The reflector in a fog light will do one of two things.....melt or be crap. Most nowadays are plastic with a chrome interior finish......and are rated for 100W tops....even the baja ones. The metal ones will have the same problem.....they are thin guage metal and the mirror surface isnt as good as commercial reflectors. Again they were designed for lower wattage bulbs. Where are you going to get a 400W MH bulb to put in one? I havent seen any 400 watters compact enough to fit in a fog light to begin with......the enclosure will most definitly melt or at the least deform.

[Gunawan W]

Daveb,
You said that your objective lens to LCD distance is 12.5" and your fresnel fl. is 8.5", supposed your reflector is parabolic type, you will get parallel light coming out from it, then it hits the fresnel, light output coming out from fresnel will be condensed to small point at fresnel focal length=8.5" in front of it. Your obj. lens is at 12.5", it's to far!
The best lens position is shorter than 8.5", try it with white hard paper as small screen, draw a circle on it with diameter=lens dia., put it in front of the fresnel, move it back and forth until the light beam covered the circle, that is the best location for your lens.
Since you already had the reflector,
first, adjust the light bulb up&down, until the light beam covered all side of the fresnel panel and
second, adjust fresnel panel position, until the light beam covered the whole objective lens.
Hope it will work. Good luck.

Woneill,
OHP fresnel can do parallel light to cone, just think it as a doublet lens, parallel beam will be focused to it's focal length.
OHP fresnel has shorter fl. than one face fresnel. (usually 12"). We can use all type of fresnel as long as it has a tight and tiny groove circle (means good quality)and a focal length we need.


Guys,
Is this the perfect reflector setup we need?
Correct me, if I'm wrong.

[woneill]

Hi Gunawan,

Your setup looks good to me!!!

For the fresnel thing, though, there is a problem using two fresnels as a DCV lens: Fresnels are usually specifically designed to have the light source and image at specific points that are fixed within a small margin. These points are called conjugates.

If you take a fresnel that is designed to have parallel light hitting one face (conjugate at infinity), and instead place a point source at its focal length, it will work, but there will be distortion and scattering. Similarly, if you take a fresnel that is expecting a point source of light at its focal length (conjugate at its focal length) and hit it with a parallel beam, you will again get a working lens, but with distortion and scattering.

The biggest problem with fresnels is internal reflection. And if they have a wide aperture and short focal length, they must be used as intended otherwise you will lose much quality.

This is the reason why they use two fresnels in an ohp, and is why the rough sides face each other - the flat side is intended to project/receive light at its focal length, and the rough side is intended to project/receive a parallel beam.

Thus in the OHP, the bottom flat side is expecting to receive diverging light from a point source bulb and converge it into a parallel beam which is fed into the rough side of the second fresnel which then converges it into a converging beam.

It WILL work as a regular DCV lens, but it is best used as intended...

See the fresnel-tech PDF for more info on fresnels:

http://www.fresneltech.com/pdf/FresnelLenses.pdf

Bill.

[daveb]

Gunwan,
Frustrated with the setup, I took the fresnel outside to the sunlight to reconfirm its focal length. As it turns out, I was finding an IMAGE of the sun (and the surrounding Gazebo) at about 8.5". I thought this is what I needed. But then I remembered all those summer afternoons frying ants on the driveway with a handheld magnifying glass, and thought (hey, shouldn't I be burning a hole in this cardboard?" So I move it a few inches close, and voila! at about 5.5", I'm burning a hole!

So, now I know that one of my problems is that my focal length of the main fresnel is really 5.5".

So I realized that if I'm using the filiment of the 400 W MH bulb as the "point source", I need to have the fresnel 5.5 inches from the filiment. But the bulb iteslf is so large I can't get that close.

So I've taken the reflector off again, put the bulb back to horizontal, and moved the fresnel down to 5.5". Still no brighter.

This is getting VERY frustrating. Now, maybe I need to cut back the reflector to accommodate the horizontal bulb, then flip over the fresnel to re-focus the light.

So here's my latest connundrum: You've suggested putting the objective lens inside the focal length of the fresnel. but I belive I'm constrained in this distance: The distnace between the object and the lens MUST be 12.5" in order to cast an image to the wall 12" away. I don't believe I have any latitude in this distance, unless I completely change my objective lens(es).

If it'll help, I'll post a couple of pix tonight of the lens head configuration.

And another question: In your diagram just posted, there's a small convex lens just beside the light bulb: if this diagram is to scale, then it's INSIDE the MH bulb. I don't understand...

dave.

[the whinner]

i found some optics software, if anyone is interesting it is at:

http://www.astrion.de/download/download.html

[thelaw]

If anyone answer this, i'd be appreciative;

In a commercial LCD projector unit (or the majority of them anyways), how does the unit know if the lamp is on? is there an optical sensor, sensitive to light? or it is measured electronically.

Im asking as I wish to retrofit a LCD projector with a cheaper metal halide.

cheers,
jon

[eebasist]

check out alternative light sources.....we've discussed this a lot in that thread.....some use a simple photocell to determine if its on....other's dont do anything at all. Projectors dont usually go into any elaborate methods for determining it.....usually the ballast is simply switched on via a relay, no complex circuitry is usually needed there.